]> www.infradead.org Git - users/jedix/linux-maple.git/commit
arm64: dts: rockchip: Fix Turing RK1 PCIe3 hang
authorSam Edwards <cfsworks@gmail.com>
Thu, 12 Sep 2024 02:50:31 +0000 (19:50 -0700)
committerHeiko Stuebner <heiko@sntech.de>
Mon, 30 Sep 2024 13:53:55 +0000 (15:53 +0200)
commit3d50680fcb31456b32cdc22581a576c388bbec4b
treee25ac5e83ccd96112134aa22a563f6e1891141cc
parent4294e32111781b3de4d73b944cbd1bc1662a9a7a
arm64: dts: rockchip: Fix Turing RK1 PCIe3 hang

The PCIe 3 PHY in the RK3588 requires a running external reference clock
for both external bus transfers and some internal PIPE operations.
Without this clock, the PCIe3 controller fails to initialize and ignores
DBI transactions indefinitely, which stalls the Linux boot process.

On most RK3588 boards, this is evidently not an issue. But on some "SoM"
designs (Turing RK1, Mixtile Core 3588E, ArmSoM AIM7, to name a few),
this clock is only provided when the CLKREQ# signal is asserted.

The PCIe 3 PHY generates the CLKREQ# signal when it knows it needs the
reference clock for proper operation. Unfortunately, the current DT for
Turing RK1 does not mux out these low-speed signals, resulting in broken
boots and potentially other issues.

This patch, following the previous one that split up the PCIe pinctrls,
resolves this problem for Turing RK1 by explicitly muxing all of the
signals needed for PCIe 2 and 3 support.

Cc: Jonathan Bennett <jbennett@incomsystems.biz>
Fixes: 2806a69f3fef ("arm64: dts: rockchip: Add Turing RK1 SoM support")
Signed-off-by: Sam Edwards <CFSworks@gmail.com>
Link: https://lore.kernel.org/r/20240912025034.180233-3-CFSworks@gmail.com
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
arch/arm64/boot/dts/rockchip/rk3588-turing-rk1.dtsi